509results about How to "Easy to track" patented technology

The invention relates to the technical field of disease follow-up visit, in particular to a health state assessment system applied to disease follow-up visit, which comprises a data acquisition module, a data preprocessing module, a health state assessment module, a personalized health management suggestion module, a dynamic health risk prediction module and a visualization module, the data acquisition module acquires multi-dimensional data of a patient and performs data fusion and integration, the health state evaluation module evaluates the health state of the patient based on multi-level feature extraction and a deep learning model, and the personalized health management suggestion module combines a health state evaluation result and historical data to obtain a personalized health management suggestion result. Customized health management suggestions are generated through a rule engine, and the dynamic health risk prediction module predicts future health risks of patients and provides real-time early warning by using a time sequence analysis model. The visualization module displays health assessment and management suggestion results through a chart and an interactive interface, and the accuracy and individuation degree of health management are improved.

The invention discloses an Internet of Things intelligent monitoring alarm method and system for water quality detection, and relates to the technical field of environmental protection monitoring, and the method comprises the steps: respectively sampling water quality electrochemical data for corresponding monitoring water area blocks based on each Internet of Things node; extracting electrochemical time sequence sensing characteristics respectively corresponding to each piece of water quality electrochemical data; extracting mode time sequence dependence features corresponding to the electrochemical time sequence sensing features based on a water quality anomaly identification model to predict water quality anomaly risks; identifying a corresponding water quality abnormity risk event type based on an abnormal event identification model; and according to the block identifier of each risk monitoring water area block and the corresponding water quality abnormity risk event type, executing water area water quality pollution alarm operation. Therefore, based on distributed water quality monitoring, intelligent risk identification and water quality pollution early warning operation of the Internet of Things nodes, water quality monitoring real-time performance, comprehensive coverage and intelligent management capability are remarkably improved.

The invention discloses a block chain-based digital right encryption management method and system, and relates to the technical field of digital right security management. According to the method, a unique identifier is generated for digital asset information through a Hash algorithm, data privacy is ensured by using asymmetric encryption, and asset uniqueness identification and information encryption protection are realized; according to the method, encrypted asset information is subjected to uplink evidence storage through an intelligent contract, node verification is carried out by adopting a consensus algorithm, tampering resistance, authenticity and transparency of the asset information are analyzed, authenticity of digital assets and safety and reliability of storage are guaranteed, legality of ownership transfer is verified through a consensus mechanism when the digital assets are transacted or transferred, and security and reliability of the digital assets are guaranteed. It is ensured that the transaction meets the preset compliance standard, the asset state on the block chain is updated in real time, and the security and compliance of the transaction are improved; the asset ownership history and the transaction path are tracked and recorded through the traceability mechanism of the block chain, and the traceability and transparency of the asset circulation process are realized.

The invention relates to the technical field of data processing, in particular to a real estate surveying and mapping data processing method and system based on intelligent data. The method comprises the following steps: acquiring multi-source real estate surveying and mapping data; performing space-time synchronization on the multi-source real estate surveying and mapping data to generate multi-source real estate surveying and mapping space alignment data; real estate space feature extraction is carried out on the multi-source real estate surveying and mapping space alignment data to obtain real estate space feature data; performing cross-domain feature fusion on the multi-source real estate surveying and mapping data based on the real estate spatial feature data to generate real estate surveying and mapping fusion spatial feature data; public surveying and mapping point confirmation is carried out on the multi-source real estate surveying and mapping data through the real estate surveying and mapping fusion spatial feature data, and real estate surveying and mapping public surveying and mapping point data are obtained. Through multi-source data space-time synchronization, feature fusion, distortion correction, high-dimensional space reconstruction, semantic tag association and chain storage, the efficiency and accuracy of real estate surveying and mapping data processing are improved.

The invention provides a target tracking method and system based on a grouped attention feature extraction network, and the method comprises the steps: obtaining a template and search features based on a template and a search image through employing the grouped attention feature extraction network; learning local areas of the template target image features and the search image features, and aggregating to obtain target information; splicing the template with the search features to obtain the input of the self-attention encoder in the asynchronous interaction stage; the fused features are combined with a prediction head for calculation, a classification regression result is obtained, and a target is predicted according to the classification regression result; training the target tracking model in combination with the large-scale data set to obtain a trained target tracking model, and performing accurate target tracking on the target by using the trained target tracking model; according to the multi-scale feature extraction network based on Transform, sequential local and global information interaction in the windows and between the windows is promoted, and the optimal balance between performance and speed is achieved.

The invention relates to the technical field of multi-round dialogue processing, in particular to a multi-round dialogue processing method based on RAG technology dynamic cue word retrieval. Comprising the following steps of question input by a user, initial retrieval and cue word generation, question splicing and answer generation, historical dialogue management and context coding, multi-round question analysis and retrieval enhancement, complex question decomposition and parallel retrieval, multi-source information fusion and generation model optimization, iterative feedback and model tuning, and exception handling and fault-tolerant mechanism. According to the method, related information is retrieved from an external knowledge base, the model can utilize wider and more specific knowledge to generate answers and better track dialogue history and context, so that answers consistent and coherent with the dialogue history are generated, the model can generate answers on a short input sequence by retrieving and splicing related cue words, and meanwhile, the model can be used for analyzing the dialogue history. Cue words are dynamically generated according to user input and retrieval results, so that the model can better adapt to different dialogue scenes and user requirements.

The invention relates to the technical field of concrete mixing, in particular to a vibrating screen mesh damage monitoring method and device based on deep learning. A vibrating screen mesh damage monitoring method based on deep learning comprises the steps that an image collecting platform is built above a single-stage discharging chute after machine-made sand is filtered through a screen mesh, and a stacked machine-made sand image in the falling process is collected; the method comprises the following steps: making images collected before and after screen damage into a dichotomy data set, inputting the dichotomy data set into a lightweight convolutional neural network for training to obtain a classification model, and deploying the classification model on an edge computer to construct an online monitoring system; and the system analyzes the collected images in real time, and sends out a screen damage alarm when judging that the number of the damaged images of the screen reaches a certain number. The state of the vibrating screen can be tracked and monitored, the damage degree of the screen can be visually mastered, and the damaged screen can be replaced in time to guarantee the production quality of machine-made sand.

The invention discloses an electronic plant air conditioner control method for setting PID parameters based on a fuzzy algorithm. The method comprises the steps that a design scheme is preliminarily determined according to plant design indexes, and a schematic diagram of a control system is planned; establishing a mathematical model; carrying out PLC control program design and configuration programming; designing a fuzzy controller; fuzzy design of input and output variables; defining an input and output membership function; establishing a fuzzy control table and rules; designing a system fuzzy reasoning algorithm; setting defuzzification output; designing a Simulink system control block diagram, and establishing a real-time control working space; and the communication between the MATLAB and the PLC is realized. According to the method, the control experience of a person is summarized into a group of conditional statements described qualitatively by utilizing a fuzzy logic function of Matlab software, and the conditional statements are quantified by utilizing a fuzzy set theory, so that the controller can accept the experience of the person, simulate an operation strategy of the person and overcome uncertain factors of the person, and the system has higher reliability.

The invention relates to the technical field of image processing, in particular to a static plantar pressure image data processing method based on an improved YOLOV5 model. Compared with traditional floor table type equipment and desktop level equipment, the static plantar pressure diagnosis system for auxiliary diagnosis of the flatfoot based on the YOLOV5 model can remarkably improve the flatfoot diagnosis effect. The improved YOLOV5 model added with the attention mechanism module can further improve the flatfoot diagnosis effect, early diagnosis and intervention are carried out on the patient, and good prognosis is achieved. Based on an advanced deep learning model and a comprehensive data analysis method, the accuracy and efficiency of flatfoot evaluation are remarkably improved through automatic auxiliary diagnosis. Through a highly automatic process, manual intervention is not needed in the whole process from data acquisition, preprocessing to diagnosis output, personal errors are reduced, and the reliability and consistency of diagnosis are improved. A non-professional user can easily use the foot sole health management system through a friendly user interface and simplified operation steps, the foot sole health management system is widely suitable for various scenes such as families, clinics and hospitals, and intelligentization and popularization of foot sole health management are achieved.

The invention discloses a document comparison method based on an improved LCS algorithm, and relates to the field of data processing, and the method comprises the steps: receiving N to-be-compared documents uploaded by a user; dividing the N documents into N groups, wherein each group comprises a main document and N-1 auxiliary documents; setting a minimum common subsequence length threshold value L0; for each group, constructing a virtual two-dimensional array, taking the character number of the main document as the row number, taking the character number of the auxiliary document as the column number, adopting a sparse matrix for storage, and only recording matched character positions; traversing the elements of the first row and the first column row by row from the elements of the last row and the last column of the virtual two-dimensional array along the main diagonal direction of the two-dimensional array, and recording the content of continuous matching characters and the starting positions and the ending positions of the continuous matching characters in the main document and the auxiliary document; in view of low comparison efficiency of tobacco purchasing files in the prior art, optimization strategies such as document grouping, multi-thread parallel comparison and clustering combination are introduced, so that the comparison efficiency is improved.

The invention provides a safety management and control system for an intelligent lock of a transformer substation, and the system comprises an intelligent wearable device which is used for monitoring the physiological data of a wearer in real time; the environment monitoring module is used for monitoring the environment risk of the working area in real time; the video monitoring module is used for monitoring the in-out dynamic state and the behavior standard of the personnel; the intelligent lock linkage management module is used for automatically locking an intelligent lock and giving an alarm according to monitoring results of the intelligent wearable device, the environment monitoring module and the video monitoring module when abnormity is monitored; the risk assessment module is used for establishing a risk assessment model and dynamically adjusting the operation authority of the intelligent lock; and the situation awareness control module is used for changing the operation authority of the intelligent lock according to the situation evaluation result. The invention further discloses a corresponding method. According to the invention, the safety and management efficiency of electric power facilities such as a transformer substation are improved.

The invention discloses a permanent magnet synchronous motor high-performance prediction control method based on a sliding-mode observer, and the method comprises the steps: S1, collecting the position and mechanical angular velocity information of a rotor, collecting the three-phase voltage and current data of a permanent magnet synchronous motor, and converting the three-phase voltage and current data into the voltage and current of a motor stator under a two-phase rotating coordinate system; s2, constructing a speed ring sliding-mode observer, estimating electrical and mechanical parameter perturbation interference and load torque interference on line, designing a rotating speed ring deadbeat prediction controller, outputting reference current at the next moment, and setting the amplitude limit of the output reference current; s3, a current loop sliding mode observer is constructed, electrical parameter perturbation interference is estimated online, a current loop prediction controller is designed in combination with the dead-beat prediction control principle, and voltage at the next moment is output; and S4, carrying out amplitude limiting modulation on the output voltage, obtaining a voltage under a static coordinate system through Park inverse transformation according to the voltage after amplitude limiting modulation, and outputting a switching signal to a three-phase inverter by combining a two-level space vector modulation module to drive the permanent magnet synchronous motor to operate.

The invention discloses a named entity recognition method and system based on a knowledge enhancement pre-training model, relates to the technical field of named entity recognition, and aims to solve the problem that an existing training model cannot perform named entity recognition more accurately. According to the method, the knowledge enhancement pre-training model is introduced, so that the model can learn more priori knowledge and semantic information, the generalization ability of unseen text data is improved, the interpretability of the model is enhanced through the detailed performance evaluation and optimization process, the working mechanism of the model is beneficially understood, and the working efficiency of the model is improved. Therefore, the model is better adjusted and optimized, the generalization ability of the optimized model is also improved, the model can better adapt to different application scenes and data sets, the generation of the monitoring log enables performance analysis and problem tracking to be more convenient, the continuous optimization of the model performance is facilitated, and the optimization efficiency is improved. A standard knowledge enhancement training model can be improved according to user feedback on a target application terminal.

The invention discloses a beam line station synchronous data acquisition device and method, and belongs to the technical field of beam line station control, and the device comprises an equipment control layer and an application layer. The equipment control layer comprises a plurality of acquisition cases, an upper computer and a time service unit, the plurality of acquisition cases are respectively connected to the upper computer, and the time service unit is respectively connected to the timing module in each acquisition case; the application layer comprises an equipment abstraction module, a process abstraction module, a queue server module, a data storage module and a data visualization module. Unified abstract packaging is carried out on beam line station equipment and an experimental process, so that the same set of software framework is universal in all experimental modes and methodology, and a unified architecture for controlling a spectroscopy, imaging and diffraction experimental process is realized; experiments can be executed autonomously according to the queue sequence, automation of the experimental process is achieved, data acquisition efficiency is improved, and then the problems that in a beam line station, multi-device synchronization efficiency is low, and synchronization signal lines are prone to interference are solved.

The embodiment of the invention relates to the field of trajectory tracking, and discloses a target detection tracking method, system and device and a storage medium, and the method comprises the steps: obtaining a plurality of image frames in a video or image sequence; performing detection by using a target detection model to obtain feature detection information of each image frame; inputting the feature detection information into a trajectory tracking model for analysis and processing, and establishing and tracking a motion trajectory of the target object; wherein the method specifically comprises the following steps of: associating the target object by utilizing the frame similarity among the plurality of bounding boxes; and establishing and tracking a motion track of the target object based on an association result. According to the method disclosed by the invention, the technical problems that the tracking object is difficult to accurately detect and the calculation time is relatively long and relatively large computing power resources need to be occupied are solved, the computing power can be reasonably distributed, the balance of high-precision detection and real-time monitoring is realized, the target object is quickly sensed in a complex shielding environment, and the target object detection and tracking effect is improved.

The invention discloses a nutrition monitoring system for an esophageal cancer patient, and aims to improve the nutritional status of the patient through real-time monitoring and personalized evaluation. The system comprises a data acquisition module, a nutrition analysis module, a personalized recommendation module and a data management module. The data acquisition module collects physiological data and diet information of a patient through the intelligent wearable device and the nutrition intake recording device. And the nutrition analysis module evaluates whether the nutrition intake of the patient reaches the standard or not by utilizing a nutrition database and a data processing analysis technology, and predicts the future nutrition demand. And the personalized recommendation module provides customized diet suggestions for the patient according to the evaluation result and the expert nutrition guide. The data management module is responsible for storing and managing the nutrition information of the patient and generating a nutrition monitoring report. The system can provide comprehensive and real-time nutrition monitoring and management for esophagus cancer patients, and is beneficial to improving the treatment effect and life quality.

The invention belongs to the field of computer vision and robots, and discloses a meta-universe dense-level instance modeling method based on neural implicit representation. Obtaining a spatial point coordinate by using the instance image, the depth image and the camera internal reference, generating an object surrounding frame, and converting the spatial point to an object coordinate system; a coordinate and plane joint coding method is adopted to carry out feature coding on the spatial points, multi-level linear interpolation is carried out to obtain object feature vectors and scene feature vectors, feature fusion is carried out, and the feature vectors are decoded to reconstruct an original image; and generating a scene Mesh and an instance scene Mesh through an SDF network by using an error optimization network of output information and a true value and a camera pose estimation result. According to the invention, tracking and map reconstruction are realized at the same time, and the local scene reconstruction precision and the camera tracking precision are improved. And an unknown area in the scene is effectively complemented based on coordinate and plane joint coding. Finally, accurate tracking precision, a high-fidelity reconstruction result and a complete semantic instance graph can be ensured.

The invention discloses an inertia damping adaptive control method, system and device, a storage medium and a product. The method comprises the following steps: constructing a virtual synchronous generator mathematical model; analyzing the power grid frequency deviation and the relationship between the power grid frequency deviation change rate and the virtual inertia variable quantity and the virtual damping variable quantity; establishing an initial fuzzy rule based on the relationship; initializing the fuzzy neural network model by using the initial fuzzy rule, and training the initialized fuzzy neural network model by using the sample data set to obtain a target model; adding the target model into the virtual synchronous generator to obtain a virtual synchronous generator based on the fuzzy neural network; and performing adaptive control on the virtual inertia coefficient and the virtual damping coefficient by using the virtual synchronous generator based on the fuzzy neural network. According to the method, the frequency deviation and the frequency deviation change rate of the power grid are used as the input of the fuzzy neural network model, and the method has better frequency response characteristics during load change and island grid connection.

The invention relates to the technical field of target tracking, and discloses a target tracking method and device, an intelligent selling cabinet and a storage medium. Performing pixel weighting on the original image data to obtain target image data; performing rotating frame labeling on the target image data to obtain labeling information of the target image data; according to the annotation information, performing feature learning on the target image data to obtain smear feature information; obtaining a target detection model according to the smear feature information; and obtaining a target tracking result according to the target detection model. According to the method, the target image data with the smear features are obtained after pixel weighting is carried out on the original data, then rotating frame labeling and feature learning are carried out on the target image data, so that a target detection model learns recognition of the smear features, and understanding of a target motion mode is enhanced. In this way, the target can be accurately detected and tracked in real time in a complex scene where the target rapidly moves.

The invention discloses a hydraulic mechanical transmission device control method based on optimal trajectory optimization, and relates to the technical field of equipment intelligent auxiliary operation and control, and the method comprises the following steps: S1, building a physical model of a hydraulic mechanical transmission device; s2, taking the shortest motion time, the most smooth trajectory and the minimum energy consumption as optimization objectives, and obtaining an optimal trajectory based on a trajectory planning algorithm combining an improved sparrow algorithm and a quintic non-uniform B-spline curve; s3, oil return control is optimized; and S4, oil inlet control is optimized, specifically, an oil inlet flow controller is optimized based on a time-varying secant obstacle Lyapunov function and an RBF neural network. According to the trajectory planning algorithm, the trajectory of the hydraulic transmission device can be optimized, oil return and oil inlet are optimized based on the optimal trajectory, and working trajectory planning and high-precision control over the hydraulic equipment executing mechanism are achieved.

The invention discloses a wear-resistant moisture-proof cable. The wear-resistant moisture-proof cable comprises a cable inner core, a supporting seat and a cable outer core, the outer side of the cable inner core is connected with a cable outer core through a supporting seat, a limiting assembly is installed on the supporting seat, the limiting assembly comprises a first through groove formed in the supporting seat, a second through groove is formed in the supporting seat, a storage groove is formed in the side end of the supporting seat, a supporting block is installed in the storage groove, and the supporting block is connected with the cable outer core. The cable inner core is limited through cooperation of the first through groove and the second through groove of the limiting assembly. According to the wear-resistant moisture-proof cable, the outer sides of the cable inner cores are protected through the supporting seats, the short circuit problem caused by too close distance between the cable inner cores is reduced, the cable damage and aging problems caused by friction between the cable inner cores are reduced, the cable inner cores are limited through the supporting seats, later use of the cable is facilitated, and the service life of the cable is prolonged. And the problems of inconvenient wiring and signal crosstalk caused by movement of the cable inner core are reduced.

The invention discloses an autonomous underwater robot system over-torque sliding mode control trajectory tracking method based on an improved adaptive law. The method comprises the following steps: establishing a six-degree-of-freedom autonomous underwater robot dynamic model; converting the six-degree-of-freedom autonomous underwater robot dynamical model, and constructing a six-degree-of-freedom autonomous underwater robot dynamical model based on lumped disturbance; designing a synchronization time estimator to estimate the lumped disturbance in the six-degree-of-freedom autonomous underwater robot kinetic model based on the lumped disturbance; and based on a sliding mode surface, a synchronization time estimator and a hyperbolic tangent function, designing a super-twist sliding mode controller based on an improved adaptive law, and realizing trajectory tracking of the autonomous underwater robot. According to the invention, buffeting of the system can be effectively suppressed while rapid convergence of the system is ensured.

The invention relates to the field of fire hoses, and provides a flame-retardant high-strength fire hose and a preparation method thereof.The fire hose sequentially comprises an outer protection layer, a flame-retardant layer, a woven layer, a reinforcing layer and a lining layer from outside to inside; the flame-retardant layer contains a surface-modified magnesium hydroxide micro-tube coated flame retardant, so that the flame-retardant effect is enhanced; the reinforcing layer contains the surface-modified bead-shaped silicon carbide nanowires, so that the mechanical property of the reinforcing layer can be remarkably improved. The five-layer composite structure and the design of multiple functional materials are adopted, comprehensive improvement of flame retardance, mechanical property, use convenience and the like is achieved, the problem that a traditional fire hose is insufficient in flame retardance and mechanical property is solved, and the fire hose has wide application value.

The invention discloses an intelligent vehicle finite time integral sliding-mode control method based on a disturbance observer, and the method comprises the steps: inputting composite disturbance into a disturbance observer model, obtaining a disturbance observer estimation value, inputting expected input linear velocity information and expected input angular velocity information, and obtaining an estimated value of the disturbance observer; the linear velocity integral sliding mode controller and the angular velocity integral sliding mode controller act to obtain position information and a vehicle running track, and the traditional sliding mode controller controls to obtain position information and a running track under a traditional sliding mode; and comparing the moving trajectory obtained by the linear velocity integral sliding mode controller and the angular velocity integral sliding mode controller with the tracking condition of the reference trajectory and the moving trajectory under the traditional sliding mode. According to the scheme, buffeting is effectively relieved, the system robustness is enhanced, the control precision is improved, the convergence speed of the system is increased, better intelligence and adaptability are achieved, better control performance and system stability are achieved, and the method is suitable for accurate path tracking and dynamic control of the intelligent vehicle.

The invention discloses a semiconductor test management system, relates to the technical field of semiconductor test, and solves the technical problems that performance detection and evaluation in different stages of wafer production are lacked, and it is difficult to detect whether a wafer has performance abnormity in the production stage in time. According to the stage performance evaluation coefficient of the semiconductor wafer of each node, a node early warning module performs stage early warning on the semiconductor wafer with stage performance abnormity, analyzes the comprehensive performance coefficient of the semiconductor wafer, and performs early warning on the semiconductor wafer which does not meet the standard; by monitoring and analyzing stage performance coefficients in real time, semiconductor wafers which do not meet the standard can be quickly identified, measures are taken in advance, and greater loss generated in the subsequent process is avoided. By detecting whether the performance of the wafer is abnormal or not in the production stage, resources can be reasonably allocated, and defective wafers can be centrally processed or re-evaluated instead of wasting time and materials on possibly failed products.

The invention discloses a time-varying parameterization nonlinear system dynamic surface control and simulation method with extended time-varying constraint, and relates to the field of nonlinear system stabilization and control, and the method comprises the steps: introducing a tangent function as nonlinear mapping in a controller design process, a nonlinear parameter system with time-varying parameters and extended full-state time-varying constraints is converted into a new unconstrained parameterized system. In addition, an improved DSC with an error compensation signal is used, and a new adaptive control technology is provided for a converted nonlinear system. The proposed design scheme avoids assuming that a virtual control coefficient and an upper limit thereof are known, and proves that all signals in a closed-loop system are in semi-global end consistency bounded. A simulation result verifies the effectiveness of the method.

The present invention provides a trough-type solar concentrating tracking bracket, wherein the adjustable lens bracket is provided with a running wheel set, and the running wheel set is driven by a motor to drive the adjustable lens bracket to move along an arc track, so that the trough-type concentrator installed on the adjustable lens bracket can rotate according to the change of the sun's position, thereby reflecting the sunlight from the most suitable direction to the solar collector tube. In this way, only the adjustable lens bracket needs to move, thereby avoiding the frequent movement of the solar collector tube, which not only reduces the maintenance and repair work of the collector tube, but also improves the high-pressure resistance of the collector tube, and at the same time greatly reduces or even avoids possible safety hazards. Therefore, the trough-type solar concentrating tracking bracket of the present invention can conveniently track the reflection of the sun, and has good practicality and market value.

The invention discloses a DHCP (Dynamic Host Configuration Protocol) security protection method and device combining an access position and third-party authentication, which effectively identify and prevent the access of illegal equipment through identity authentication and an access position binding mechanism and reduce the possibility of implementing network attacks by fake identity or disguised equipment. And through real-time abnormity monitoring and alarming, the difficulty of tracing after the event is reduced. According to the method, the configuration of the client does not need to be changed, an operator can quickly deploy and implement, and the existing network architecture does not need to be modified on a large scale. Through the automatic authentication and binding process of the system, the operation burden of an administrator is relieved, and the overall management efficiency is improved. According to the embodiment of the invention, through detailed recording and unique binding of each access device, an accurate device access history can be provided, and tracking and responsibility investigation can be conveniently carried out when a security event occurs. The scheme only depends on the existing network facilities, no extra hardware equipment needs to be purchased for implementing the scheme, and the implementation cost is reduced.

The invention provides a collaborative robot man-machine interaction and anti-collision control method and device based on admittance, and relates to the technical field of robot intelligent control. The method comprises the following steps: acquiring control input force; according to the control input force, constructing a relationship between a state variable and control input in the man-machine interaction system; according to the relation, an expected admittance model and a soft saturation function are adopted, and a reference trajectory of the robot end effector is generated; according to the reference trajectory, an Actor-Critic algorithm is adopted, and a reinforcement learning neural network controller based on admittance control is designed; according to the reinforcement learning neural network controller based on admittance control, a control strategy in the man-machine interaction system is adjusted, and an anti-collision strategy in the man-machine interaction system is obtained. According to the invention, the safety and stability of the man-machine interaction system are improved.

The embodiment of the invention discloses a tractor farm tool unit control method and device, electronic equipment and a storage medium. The method comprises the steps that a state space expression of a control system of a tractor farm tool unit is established; the state variables of the control system comprise the tractor course deviation, the farm tool course deviation and the farm tool transverse deviation of the tractor farm tool unit relative to the target operation path, and the control variable of the control system is the front wheel rotation angle of the tractor; acquiring a state feedback controller of the control system by adopting a linear quadratic regulator based on the state space expression; based on the real-time tractor course deviation, the real-time farm tool course deviation and the real-time farm tool transverse deviation, a real-time target front wheel rotation angle is determined through a state feedback controller; and performing real-time regulation and control on the tractor front wheel rotation angle of the tractor farm tool unit based on the real-time target front wheel rotation angle. According to the embodiment of the invention, the farm tool can better track the target operation path.